24. PETROGRAPHY AND GEOCHEMISTRY OF RED SEA DOLOMITE
Peter R. Supko,1 Peter Stoffers,2 and Tyler B. Copied
INTRODUCTION in the Site Report section of this volume and are further discussed by Stoffers and Ross (this volume). Generalized Examination of piston cores and, particularly, sediments stratigraphic columns with unit correlations are shown in collected by the Deep Sea Drilling Project has shown that Figure 2 and unit boundary depths, in terms of both meters authigenic dolomite occurs in a wide range of sedimentary subbottom and core numbers, are given in Table 1. associations. While the trace to a small percentage of Unit I—This unit is basically a biogenic calcareous ooze dolomite often encountered in pelagic carbonate sequences with varying admixtures of detrital silt and clay. may be a result of authigenesis under "normal" deep-sea Nannofossil platelets and foraminifera commonly make up conditions, the few studies to date seem to indicate that the calcareous fraction, with a significant (10%-30%) dolomite is abundant only in sediments presumably formed addition of micarb4 which, in the Red Sea, is thought to under conditions other than normal open marine. For mostly represent diagenetically altered coccolith material. example, certain dolomite sequences have been ascribed to Dolomite is present as rhombs, from a trace to a maximum magnesium enrichment associated with igneous activity of 10 percent. Detrital silt and clay constitute 20 to (rarely) (Bonatti, 1966; Riedel et al., 1961). In other cases 50 percent of unit I sediment at Sites 225 and 227. At Site dolomites are thought to have formed in association with 228 the overall detrital contribution is higher, including hypersaline brines. A DSDP core (Site 12) from the African sands of complex mineralogy as a major constituent, continental margin contained abundant dolomite and reflecting the proximity of an active Plio-Pleistocene Sudan palygorskite. Peterson, Edgar, et al. (1970) feel that Delta (see Site 228 chapter). hypersaline brines, formed in near-shore lagoons, moved Unit I sediments are generally well burrowed, but down through the deep-water sediment sequence by a occasional layering is seen as color variations, with the process of refluxion. In other cases, hypersalinity is colors being indicative of redox potential. White, brown, associated with desiccation. Miocene sediment sequences and reddish colors indicate generally oxidizing conditions, drilled in the Mediterranean contain abundant dolomite whereas the darker colors (grays, greens) indicate reducing associated with halite and anhydrite. Ryan, Hsü, et al. conditions. Darker layers contain pyrite and more abundant (1973) postulate that these sediments formed during a organic matter. "salinity crisis." Another compelling association is that of Lithified carbonate layers, sometimes rich in pteropods, dolomite and inferred reducing conditions noted in cores occur in the upper sections of unit I at Sites 225 and 227. recovered by DSDP from the African continental margin The cement, which is most often composed of aragonite and the Cariaco Trench of the Caribbean Sea (Hayes, Pimm, and sometimes of high magnesium calcite, reflects periods et al., 1973; Edgar, Saunders, et al., 1973). Whether organic at which these minerals were inorganically precipitated, matter itself is a big factor, or whether hypersalinity perhaps in conjunction with eustatic fluctuations of sea associated with environmental extremes in these inferred barred basins is controlling, is still a problem. level (or tectonic movements of a sill area). Unit II—Sediments of unit II are basically the same as Dolomite was found to constitute a significant part of those of the overlying unit except for a higher terrigenous portions of the lithostratigraphic sequences at three sites content. As in unit I, there are light gray and dark gray drilled in the Red Sea. These dolomites were studied layers, the latter enriched in pyrite and organic matter. petrographically and geochemically to determine their Unit III—This unit is a dark gray to black semilithified nature, to associate them with the depositional environ- dolomitic silty clay stone. Dolomite commonly makes up 20 ment as deduced by other means, and to see if they percent of the total, locally attaining 80 percent. The themselves could add to our knowledge of that claystone contains abundant analcite in the clay fraction, environment. 5-10 percent pyrite, and is rich (up to 6%) in organic carbon. Millerite (MS) is also present. LITHOSTRATIGRAPHY Unit IV—Unit IV is an evaporite sequence of alternating Sites 225, 227, and 228 were drilled (and continuously anhydrite and halite. The anhydrite is both of laminar and cored) in the central Red Sea Main Trough, 225 and 227 nodular texture; the halite is light gray to clear. The being to the east of the Axial Trough and 228 to the west evaporites are described in detail by Stoffers and Kuhn (this (Figure 1). The same four lithostratigraphic units were volume). Dolomite and pyrite occur in abundance as matrix recognized at all three sites. These units are fully described between anhydrite laminae and nodules and as an important constituent of black claystones which are layered within the evaporite sequence. Scripps Institution of Oceanography, La Jolla, California. Laboratorium fur Sedimentforschung, Universita^ Heidelberg, Germany. Institute of Geophysics and Planetary Physics, University of Micarb is a term used to refer to very fine carbonate particles of California, Riverside, California. uncertain origin.
867 P. SUPKO, P. STOFFERS, T. CAPLEN
RED SEA
2 5'
EGYPT
HOT BRINE AREA
20' SUDAN SAUDI ARABIA
15'
ETHIOPIA
35° 40° Figure 1. Base map showing the locations of Sites 225, 227, and 228 in the Red Sea.
868 SITE S I TE 225 227 O-i
>LITHIFIED LAYERS
CALCAREOUS (NANNO) OOZE and CHALK, variable DETRITAL admixture 100-
CALCAREOUS SILT (STONE) and CLAY (STONE)
200-
CLAYSTONE, -ich in ORGANIC MATERIAL, DOLOMITE, PYRITE EVAPORITES, with interbeds of PYRITIC UU=1 CLAYEY SILT to SAND DOLOMITIC CLAYSTONE DOLOMITE 300-1 ANHYDRITE
> ö o r oo i Figure 2. Correlation of generalized lithostratigraphic sections at Sites 225, P. SUPKO, P. STOFFERS, T. COPLEN
TABLE 1 coccolith plates which, by their relatively poor preserva- Unit Boundaries tion, appear to be diagenetically altering. Significant are the two round nanno(?) plates attached to the rhomb in the lower right corner of the illustration. This is taken to Sediment indicate recrystallization in process. Another such rhomb Unit Site 225 Site 227 Site 228 (same sample) is shown in Figure 4. Note here the Unit I 0-112 m 0-131 m 0-195 m (Cores 1-17) (Cores 1-17) (Cores 1-24) Unit II 112-167 m 131-194 m 195-250 m (Cores 18-22) (Cores 18-25) (Cores 24-30) Unit III 167-176 m 194-226 m 250-287 m (Core 23) (Cores 26-29) (Cores 30-35) Unit IV 176- ? m 226- 7 m 287- ? m (Cores 24-29) (Cores 30-45) (Cores 35-39)
Deposition of the units at the three sites has been roughly synchronous (Figure 2), with evaporite deposition ceasing at the close of the Miocene. Unit III, the dolomitic claystone, was deposited in roughly the Early Pliocene and the biogenic deposit is subsequent to this. Since units I and II differ only in the relative amount of detrital material admixed, no further distinction will be made between them.
PETROGRAPHY OF DOLOMITES Petrographic characteristics of the dolomites were Figure 4. A unit 1 dolomite rhomb lying in a matrix of studied in thin section under the petrographic microscope partially recrystallized coccoliths. Note the sliver-like (Peter Stoffers—Heidelberg) and by scanning electron particles apparently crystallizing onto the top of the microscopy (Peter R. Supko—Scripps) of slurried and fresh rhomb. Slurried sample (SEM) 225-16-2. break samples. In general, dolomites of units I and II are relatively large recrystallized coccoliths to the upper and lower right of the euhedral rhombs, while those of units II and IV are fine rhomb and the thin sliver-like particles apparently grained and anhedral. crystallizing onto the top face of the rhomb. A typical unit I rhomb is seen in Figure 3. It is euhedral Dolomites of unit II, the basal and more detrital-rich and has a crystal edge about 6µ long and is lying among section of the ooze facies, are similar and, if anything, more perfectly euhedral. Figure 5 shows dolomite euhedra lying in a matrix of clay and recrystallized carbonate. Figure 6 shows a cluster of rhombs with mutually interfering crystal faces, clearly indicating in situ growth. At higher magnification, several crystals exhibit a "spalling" or "peeling off" of crystal faces (Figure 7, note crystals in upper right center and lower left). Even if this occurred during sample preparation (this is a fresh break sample), it might still be indicative of lattice inhomogeneities. In contrast, SEM observation detected no dolomite euhedra in the half dozen unit III and IV samples studied. Figure 8 depicts a sample from the center of unit III at Site 227. Although X-ray diffraction analysis shows this sample to contain 25 percent dolomite and 75 percent detrital material, repeated scanning showed nothing other than intergrown fine anhedral crystals. An SEM photograph from the center of unit IV at this same site (227) shows a general mosaic of intergrown anhedra (Figure 8). This sample represents a dolomite-pyrite seam between anhy- drite nodules. In unit IV, dolomite is largely associated with organic Figure 3. Typical euhedral dolomite rhomb from unit I. matter and may occur as scattered anhedral crystals within Note the two round nanno (?) plates which appear to be anhydrite (Figure 9) or may be a replacement of an organic recrystallizing onto the rhomb. Slurried sample (SEM) structure, in this case an oncolite (Figure 10, see also 225-16-2. photomicrographs in Stoffers and Kuhn, this volume). Note
870 PETROGRAPHY AND GEOCHEMISTRY OF RED SEA DOLOMITE
Figure 5. Typical unit II dolomite euhedra lying in a Figure 7. Higher magnification of Figure 6 showing several matrix of clay and recrystallized carbonate. Fresh break crystals exhibiting a "spoiling" or "peeling off" of sample (SEM) 225-21-1. crystal faces.
Figure 8. Typical SEM photomicrography of unit III fades. Although X-ray diffraction analysis indicates this sample contains 25 percent dolomite, no dolomite euhedra are Figure 6. Same sample as Figure 5, here showing a cluster seen. Fresh break sample (SEM) 227-28-2. of rhombs with mutually interfering crystal faces. calcium. For example, Figure 12 shows relative (112) peak in Figure 10 the fine, anhedral nature of the dolomite positions of samples from three cores from different depths crystals and their darkness due to organic inclusions. The subbottom (i.e., different stratigraphic positions) at Site vug in the lower left center is partly filled with anhydrite. 225. The (112) peak of the sample closest to the top of the Dolomite, associated again with pyrite and organic material, section (Core 21) is at a low 20 angle, indicating significant may form matrix separating anhydrite nodules or discrete excess calcium, whereas the (112) peaks of samples from bands in laminar anhydrite (Figure 11). Cores 25 and 29 are shifted to the higher 20 values indicating progressively less excess calcium with depth. GEOCHEMISTRY Figure 13 contains 29 data points from the three sites. Excess mol % CaCU3 is plotted as a function of depth in Excess Calcium meters subbottom. Core numbers and unit boundaries are Twenty-nine dolomite samples were analyzed for excess shown. The trend of decreasing excess calcium in the calcium by the X-ray diffraction peak shift method dolomite lattice with depth in section (i.e., as the evaporite (Goldsmith et al., 1955). Diadochic substitution of Ca for sequence is approached) is apparent at all three sites but is Mg in the Mg cation planes in the dolomite lattice causes best shown at Site 225. Excess calcium decreases with shift of the (112) diffraction peak to lower 20 values. The depth in the section, finally achieving stoichiometric degree of peak shift is a function of the amount of excess proportions (CasoMgso) in the unit IV evaporite sequence.
871 P. SUPKO, P. STOFFERS, T. COPLEN
o< o< CO >o SITE 225 CO o DOLOMITE 00 oi CM (112) PEAKS o< θ Q)CORE 21 (149-158m) ; (CA57 Mg43) I f ' 1 l (2) CORE 25(185-194m) I (CA Mg ) I 53 47 I Q) CORE 29(221-230m)
(CA50 Mg50) Figure 9. Unit IV dolomite, occurring as scattered anhedral crystals within anhydrite. The dark color is imparted to the dolomite by finely dispersed organic matter. Thin section photomicrography, crossed nicols, 228-36, CC.
1 1 1 1 »o r— O u•> 1—' CO ö CO CS CO CO CN °2θ Figure 12. Relative shift of the (112) dolomite peak for three samples at different subbottom depths, Site 225.
Ordering Figure 10. Organic-rich dolomite replacing what appears to As the stoichiometric condition is approached, calcium be an oncolite. This section photomicrography, plane and magnesium cations become better distributed into light, 225-26-1. discrete basal cation planes. This gives rise to a series of X-ray reflections (ordering peaks) from these planes. Degree of ordering is measured by the relative height of an ordering peak to another peak not directly associated with these alternating cation planes. There is a slight trend toward increased ordering going from unit III to unit IV, but it is not well enough shown by the relatively few samples run over this portion of the X-ray spectrum to warrant illustration here. The dolomites in the evaporite sequence are well ordered. Figure 14 shows the relative peak heights of three dolomites associated with the evaporites. The ordering peak at 35.35°20, normally absent or just a low bulge in poorly ordered dolomites, is here quite sharp. Measured peak height ratios relative to the nonordering peak at 37.25°2ö (R values in Figure 14) are all high when the scale used by Fiichtbauer and Goldschmidt (1965) in their study of dolomites of the Zechstein is applied. Figure 11. Unit IV dolomite, associated with pyrite and organic material, separating bands of laminar anhydrite. Isotopes Thin-section photomicrograph, crossed nicols, 227-37, Eleven samples were analyzed for δθ18 and öC*3 on a CC. double-focusing isotope ratio mass spectrometer (McKinney
872 PETROGRAPHY AND GEOCHEMISTRY OF RED SEA DOLOMITE
CαCOß excess of Red Sea dolomites «
SITE 225 0 194H O »< SITE 228 u i u 26 203- 245- 27 30 212- 149 254 28 21 31 221- 158 UNIT III 263- 29 L 32 22 • 226-f UNIT II ^ > UNIT IV 268- 167 - 33 23 235 277- 31 34 176- UNIT III 244- 24 UNIT IV 286 32 UNIT III 185- 35 253- "UN7T"IV" 25 295- 194 33 36 26 262- 304H 34 203- > 37 271H 27 313H 212- 38 1 ALSO 322 28 39 > JMAGNESITE 325- 221-
12345678 12345678 12345678 MOL %
Figure 13. Excess mol% CaCO3 as a function of depth at Sites 225, 227, and 228 as measured by the peak shift method.
et al., 1950; Coplen, 1973) at the University of California, versus nonequüibrium conditions, isotopic composition of Riverside laboratory using the method of McCrea (1950). precursor sediments, etc. Notwithstanding these many Of these, eight contained both calcite and dolomite, three complicated and interrelated factors, most of which are contained dolomite only. Dolomite was separated from unknown, the Red Sea data allow some insights. calcite by their differential ractions with 100 percent Very striking is the extreme range of isotopic values for phosphoric acid—at the end of 1 hour's reaction, the gas δθl8 and δC13 for both calcite and dolomite. The and in extracted was assumed to be primarily calcite derived; the dolomites range in δO18 from +2.47 to -6.30°/oo 3 gas derived from 1 and 4 hours was pumped away. The gas δCl from +2.74 to -5.69°/oO; the calcites range inδC lS 13 derived between 4 hours and 7 days was assumed to be from +0.78 to -8.83°/oo and in δC from +1.18 to derived primarily from the dolomite. The data are given in -7.25%o. Also the dolomites from within unit IV 18 13 18 Table 2 and are shown as a plot of δθ versus δC in themselves have a wide range of δθ values (-0.70°/oo for Figure 15. All data are relative to the Chicago PDB-1 225-27-2, -2.89°/oo for 227-36-2, -4.42%o for 227-30-1, belemnite standard and have been corrected for the and -6.30%o for 227-29-5; Table 2, Figure 16). These phosphoric acid fractionation factor (Sharma and Clayton, ranges are interpreted as being environmentally significant 1965). and are considered in the Discussion section following. Stable isotope data must be used with extreme care as In addition to the above, the Red Sea isotopic data supporting data in arguments bearing on dolomitization and suggest trends which workers may find helpful in carbonate diagenesis because of their dependence on a considering the overall problem of utilization of stable number of factors, such as: temperature at time of isotope data in carbonate studies. We look upon these formation, isotopic composition of reservoir water at time trends only as guides to needed future research and caution of formation, reequilibration with a reservoir different from that any statements made here are more food for thought that at time of formation, precipitation under equilibrium than they are hypothesis.
873 P. SUPKO, P. STOFFERS, T. COPLEN
ORDERING PEAKS OF TABLE 2 RED SEA DOLOMITES Stable Isotope Analyses of Selected Red Sea Carbonate Samples
Sediment Sample Unit Mineral δCl3 5018 ΔO18 Λ 227-3-1, I Calcite +0.47 -0.30 27 cm Dolomite +0.85 +1.38 +1.68 \ 227-5-2, I Calcite +0.78 -0.93 R--0.81 \ 120 cm Dolomite +2.74 -0.40 +0.53 ^ / \ 225-9-6, I Calcite +1.18 +0.78 ^– /225-27- V 2 132 cm Dolomite +1.75 +2.47 +1.69 (UNIT III) 228-24-3, HI Calcite -1.72 -1.99 28 cm Dolomite -2.93 +1.99 +3.98 A 225-21-1, II Calcite -3.16 -4.12 125 cm Dolomite -5.69 -0.10 +4.02 / 228-30-5, II-III Calcite -2.55 -4.12 R=0.77 \ 37 cm Dolomite -3.23 -0.30 +3.82 227-29-5, III-IV Calcite -0.60 -3.05 Dolomite +1.20 -6.30 -3.25 227-29-5 227-30-1, IV Dolomite -5.01 -4.42 N/A (UNIT III-IV) 135 cm 227-27-2, IV Dolomite -4.08 -0.70 N/A 60-70 cm 226-36-2, IV Dolomite -0.98 -2.89 N/A i a 155 cm R=0.62 225-28-1 IV Calcite -7.25 -8.83 \ A Dolomite -4.04 -4.14 +4.69 328-37-2 (UNIT IV) STABLE ISOTOPES 37' 3*5- RED SEA CARBONATES Figure 14. Relative heights of ordering peaks (35.25° 2θ) to nonordering peaks (37.25° 2θ) of three samples of dolomite asociated with the evaporites. R values apply +3-,
to the scale of Füchtbauer and Goldschmidt (1964). +2
First, there is a general tendency toward more negative +1- oxygen values with depth (Figure 16). This could be explained by one or both of the following: O
1) Red Sea carbonates lower in the section formed at -1- progressively high temperatures. c 2) Red Sea carbonates lower in the section formed in waters depleted in O18. a The depletion of 018 with depth in carbonate sediment has also been observed by Anderson and Schneidermann Q.-3 (1973) and by Coplen and Schlanger (1973) in studies of S-4 recrystallization of deep-sea carbonate oozes. In these cO~5 deep-sea studies, the cause of this decrease in 018, extending back to Jurassic time, is not well understood. -6 An interesting observation is the enrichment in 0*8 of dolomite relative to calcite when both mineral species occur -7 5 8 18 in the same sample. The Δ0D -C (SO dolomite value -β minus the δθ18 calcite value) range6 is from +0.53 to -9 +4.69%o, averaging +2.91%o. This may represent the 7 ^6 I3 ^4 ^3 ^2 -1 0 +1 +2 +3 6C" rel PDB in‰ 5 With the exception of the mineral pair in Sample 227-29-5, ©DOLOMITE CALCITE 18 which shows a O of -3.5°/oo. It is tempting to suspect that the analytical values may have been erroneously recorded in the logging Figure 15. Isotopic ratios of Red Sea dolomites and procedure. The sample will be rerun. co-existing calcites, expressed in %o notation relative to 6Excluding Sample 227-29-5. the Chicago belemnite standard.
874 PETROGRAPHY AND GEOCHEMISTRY OF RED SEA DOLOMITE 8018(%o)
-9 -8 -7 -3 -2 -1 +1 +2 + 3 +4 _l_ L I
227-5 227-3(1) 227-3 ® 227-5(1) o _ 225-9 225-9 • (I) (I)
100-
25-21 (» 225-21 ® (π) 228-24 228-24 #α-π) 227-27 200- 225-28 225-28 227-30. >
228-30 228-30 s (π-m)
227-36 300- KEY: CALCITE DOLOMITE ®
400-1 Figure 16. Values o, o/i?ec? Sea dolomites and calcites plotted versus subbottom depth. fractionation occurring when the two minerals coprecipi- Clayton (1966) to have higher fractionations (in their case, tate in chemical equilibrium. Early work (Clayton and 4-5%o) than those of the calcium-rich and poorly ordered Epstein, 1958; Engel et al., 1958) indicated that an protodolomites of Fritz and Smith (1970, ΔC>l8 of equilibrium isotopic fractionation factor between dolomite 34°/oo) which are more analogous perhaps to the unit I and calcite should be greater than one, and experiments by and II Red Sea dolomites. Second, the observed trend with Northrop and Clayton (1966) determined it should be depth may be a result of cross-contamination during sample about 1.004 to 1.007 at 25°C (i.e., approximately 4-7%o) acidification. In calcite-dolomite mixtures containing for calcite and stoichiometric dolomite. Certain field well-ordered stoichiometric dolomite, mutual contamina- studies (e.g., Degens and Epstein, 1964), however, found tion may be minimal (Northrop and Clayton, 1966), but co-occurring calcite and dolomite with essentially the same the more calcium-rich the dolomite, the more easily it tends δθ18 values. Clayton et al. (1968) observed that dolomite to react with the acid, thus the possibility of contaminating was enriched by several per mil relative to coexisting calcite the "calcite" phase exists. in Deep Springs Lake, California. Many authors have since used isotope data variously to support different hypotheses of dolomitization, dependent upon their belief or nonbelief DISCUSSION in a fractionation factor. The question has still not been A brief and general outline of the sedimentation history resolved to the satisfaction of isotope geochemists, but of the Red Sea, as deduced from Leg 23 cores, will help to recently Fritz and Smith (1970) have shown that calcite put the question of the origin of the dolomite into and protodolomite (containing several mol % excess perspective. For detail, see Stoffers and Ross (this volume). CaCC>3) precipitated in the laboratory show a ΔOr>c °f Sedimentary unit IV consists of Late Miocene halite and 34%o. anhydrite, with local interbeds of pyrite- and dolomite-rich 18 Another interesting trend indicates the ΔO to increase black organic shales. Studies by Leg 23 scientists, and by with depth (Figure 17). This cannot itself be a temperature Leg 13 scientists on correlatible Late Miocene evaporites in effect on a fractionation factor, since fractionation should the Mediterranean Sea, indicate numerous criteria for decrease with increasing temperature. Two possibilities may shallow-water deposition, including textures, sedimentary be considered. First, the dolomites deeper in section (units structures, presence of stromatolites and oncolites, II and IV) more closely approach stoichiometric propor- geochemical evidence, etc. The depositional environment tions and are better ordered, as shown in earlier sections. was sabkha-like, for the most part hypersaline, but marked Such are the type of dolomites shown by Northrop and by occasional incursions of fresh continental waters.
875 P. SUPKO, P. STOFFERS, T. COPLEN
18. 0 ir 1965). Close physical and temporal proximity of unit III +1 +2 +3 + 4 +s and IV dolomites to anhydrite and halite indicates that _, | such conditions existed at the time of deposition of these units. Magnesium enrichment is indicated by the presence
(I) of authigenic palygorskite (Stoffers and Ross, this volume). (I) On the other hand, the association of the dolomite with organic matter is also compelling. Recent studies of algal (I) stromatolites (e.g., Gebelein, 1971) indicate certain species of mat-forming algae, through increase in Mg/Ca ratio, may foster the precipitation of high magnesian calcite, which in 100- turn may later recrystallize as dolomite. Laminated dolomite-rich layers in the Red Sea cores have been interpreted by Stoffers and Kuhn (this volume) as representing stromatolites and oncolites. Davies and Supko
m 225-21 (1973) note a relationship between dolomite occurrence and inferred reducing conditions in Deep Sea Drilling cores from Legs 14 and 15. Gieskes (personal communication) notes that, although dolomite at Site 147 (Cariaco Trench, A 228-24 200- w
228-30 occasional fluctuations of sill depth and/or eustatic sea level m might result in hypersalinity. On the other hand, Donnelly (personal communication) notes the association of dolomite with sulfate reduction in the Site 147 (Cariaco
300 J Trench) Caribbean Sea cores. Lippmann (1968), based upon analogy with norsethite (Ca/Mg [Cθ3]2), which is 1S Figure 17. AO of Red Sea dolomite-calcite pairs plotted similar chemically and structurally to dolomite, states that as a function of depth subbottom. early diagenetic dolomite should form in the presence of Mg++, given sufficiently high alkalinity. Alkalinity may be The evaporites are overlain by dolomitic silty clay stone, raised by bacterial reduction of sulfate. Spotts and rich in organic matter and pyrite. These deposits were Silverman (1966) describe organic dolomite from Point apparently laid down in an Early Pliocene semirestricted Fermin, California. basin. These euxinic deposits are overlain by the The dolomites of units I and II differ from those of units progressively more open marine deposits of units II and I. III and IV in being larger and more euhedral. Stoffers and The dolomite associated with the evaporites of unit IV Ross (this volume) have noted that these dolomites are was seen to be very finely crystalline, generally indicative of more abundant in sedimentary zones having higher detrital early diagenetic or primary formation, and, in the samples content and conclude that they may therefore represent examined, anhedral. Similar dolomite interbedded in the detrital dolomite. On the other hand, the euhedrality, Messinian evaporites of the Mediterranean is also fine evidence of incipient recrystallization, and mutually grained, but appears more euhedral under SEM (Nesteroff, interfering rhombs indicate that at least a portion of the 1972). In both the Mediterranean and Red seas, the dolomite is authigenic. The controlling factor for dolomites are associated with organic matter and pyrite. dolomitization here is unknown. Although periods of The Mediterranean samples contain relict biogenic material, hypersalinity are inferred to have caused precipitation of nannofossils, and foraminifera, suggesting the dolomites are aragonite and high magnesian calcite in the surface early diagenetic replacements of marine oozes. The oozes sediment layers of the Red Sea (unit I) (Milliman et al., would have been deposited during brief marine incursions 1969), dolomite was not found. Whether or not reduction between periods of desiccation in a sabkha environment, of organic matter may be important here is as yet with attendant deposition of halite and anhydrite. The impossible to say. fine-grained, anhedral dolomites of unit III were probably The increase in mol % MgCθ3 and degree of ordering also formed as penecontemporaneous replacements of with depth in section (i.e., with proximity to the marine carbonates under conditions only slightly less evaporites) seem to indicate the importance of hyper- environmentally severe. salinity for dolomite formation. Goldsmith and Graf (1958) The question of the triggering mechanism for this early have postulated that higher salinities favor the precipitation diagenetic dolomitization remains. Fine-grained Holocene of well-ordered dolomites. Füchtbauer and Goldschmidt dolomite muds supposedly form by early diagenetic (1965) reported that dolomites formed at higher salinities replacement of calcium carbonate in a supratidal or sabkha contain less excess CaC03 in the lattice, based upon field environment under the influence of hypersaline brines of studies of carbonates from the Lower Tertiary of Libya and high Mg/Ca ratio, this ratio increased via precipitation of Upper Permian and Upper Jurassic of northern Germany. calcium sulfate (e.g., Deffeyes et al., 1965; Bring et al. Similarly, Marschner (1968), studying the Triassic Keuper
876 PETROGRAPHY AND GEOCHEMISTRY OF RED SEA DOLOMITE
Formation of northwest Germany, concludes that the grained and anhedral. They are interpreted as early Mg/Ca ratio of dolomite is dependent upon degree of diagenetic (penecontemporaneous) recrystallization prod- hypersalinity at the time of formation and that ucts of original carbonate deposits, probably marine stoichiometric dolomite is formed only under conditions of biogenic oozes. They formed in a hyper saline environment, hypersalinity. The well-defined trend toward decrease in in association with organic matter. Hypersaline brines with excess CaCC>} with depth in section (i.e., approach toward high Mg/Ca ratios brought about by calcium sulfate evaporites) at these three Red Sea sites seems to confirm precipitation were probably the cause of dolomitization. the results of these earlier studies. Alkalinity increase by sulfate-reducing bacteria may have The wide range of δθ18 and δC13 values for the played a role in dolomitization as well. dolomites of units III and IV (Table 1, Figure 15) is 2. Dolomites of units I and II are larger and more interpreted as indicating conditions of environmental euhedral. While some of the dolomite may be detrital, extremes. In a sabkha-like or desert environment, the interfering crystal boundaries, euhedrality, and inferred reservoir supplying oxygen anions for the recrystallization recrystallization in process indicates that at least some is process might at times be enriched in O18 through authigenic. The mechanism is unknown. evaporation, whereas at other times influx of fresh waters 3. A trend toward increased ordering and, more might significantly lower the O18 content. Similarly, fresh strongly, decrease in excess CaCθ3 in dolomite as the waters, draining through soil horizons, might occasionally evaporite sequence is approaced substantiates work by decrease the C13/C12 ratio below that associated with others indicating that increased salinity favors the evaporative conditions. precipitation of better ordered and more stoichiometric Fontes et al. (1973) and Lloyd and Hsü (1973) found dolomite. wide ranges of isotopic values for dolomite samples from 4. Oxygen isotope analyses of carbonates suggest that the evaporite sequences of the Mediterranean Sea and recrystallization of carbonates has occurred in units I and invoked rapid changes of isotopic composition of a shallow IL The wide variation in O18 content in units III and IV body of brine to explain them. A similar conclusion was suggests that the carbonates have not recrystallized or reached earlier by Parry et al. (1970) in their studies of isotopically equilibrated over a scale of tens of meters at a West Texas lake carbonates, again formed under conditions minimum. The enrichment of dolomite in Ol8 relative to of climatic extremes. Deuterium/hydrogen ratios (Friedman coexisting calcite by a few per mil is a trend which should and Hardcastle, this volume) of pore waters support be further explored. meteoric water dilution of Miocene Red Sea brines. Finally, the general trend toward more negative δθ18 values for the calcites and dolomites in the Red Sea samples ACKNOWLEDGMENTS may be a temperature effect. The more negative values with The authors thank Drs. Melvin N. A. Peterson and Joris depth in units I and II may reflect fairly recent (and even Gieskes, Scripps Institution of Oceanography, for critically present) recrystallization at progressively higher tempera- reviewing the manuscript. Since certain suggestions were tures with depth, an effect of geothermal gradient. not followed, errors of omission or commission are solely Assuming this is the case, one can estimate from the points the authors'. The senior author's research was supported by for calcite in units I and II in Figure 16 a slope of0.01%o the U. S. National Science Foundation, Contract C-482. Dr. per meter. If we assume recrystallization of calcite with an Stoffers was partially supported by a NATO postdoctoral infinite source of water, the "isotope geothermal gradient" fellowship. would be approximately 4°C per 100 meters. The dolomites of units HI and IV, thought to be early diagenetic (penecontemporaneous) cannot be explained by a recrystallization mechanism. The high temperature condi- REFERENCES tions, as suggested by the very light O18 values of the Anderson, T. F. and Schneidermann, N., 1973. Stable carbonates, may have been brought about in part by isotope relations in pelagic limestones from the central atmospheric (i.e., near surface) conditions, partly due to Caribbean, Leg 15. In Edgar, N. T., Saunders, J. B., et extreme insolution and partly to adiabatic effect in the al., Initial Reports of the Deep Sea Drilling Project, presumed deep basin. Volume 15: Washington (U. S. Government Printing Alternatively, the decrease in δθ18 with depth may not Office), p. 795-803. be a temperature factor at all. It may represent a Bonatti, E., 1966. Deep-sea authigenic calcite and progressively larger degree of incorporation of isotopically dolomite: Science, v. 153, p. 54-57. light oxygen from bicarbonate produced as a result of Clayton, R. N. and Epstein, S., 1958. The relationship sulfate reduction. Or, it may represent a progressive between 0*8/016 ratios in coexisting quartz, carbonate increase in dilution by meteoric water with depth. Much and iron oxides from various geological deposits: J. Geol., v. 66, p. 352-373. additional work is clearly needed in this field. For a Clayton, R. N., Jones, B. F., and Berner, R. A., 1968. discussion of the complex factors involved, see Lawrence Isotope studies of dolomite formation under sedi- (in press). mentary conditions: Geochim. Cosmochim. Acta, v. 32, CONCLUSIONS p. 415-432. Coplen, T. B., 1973. A double-focusing double-collecting 1. Dolomites within and immediately above the mass spectometer for light stable isotope ratio analysis: Miocene evaporite series in the Red Sea cores are fine Int. J. Mass Spectrom. Ion Phys., v. 11, p. 37-40.
877 P. SUPKO, P. STOFFERS, T. COPLEN
Coplen, T. B. and Schlanger, S. O., 1973. Oxygen and carbonates, and silicates from the Venezuela Basin (Site carbon isotope studies of carbonate sediments from Site 149) and the Aves Rise (Site 148). In Heezen, B. C, 167, Magellan Rise, Leg 17. In Winterer, E. L., Ewing, MacGregor, I.G., et al., Initial Reports of the Deep Sea J. I., et al., Initial Reports of the Deep Sea Drilling Drilling Project, Volume 20: Washington (U. S. Govern- Project, Volume 17: Washington (U. S. Government ment Printing Office). Printing Office), p. 505-510. Lippmann, F., 1968. Syntheses of CaMg (COßh (Norsethi- Davies, T. A. and Supko, P. R., 1973. Oceanic sediments te) at 20° C and the formation of dolomite in sediments. and their diagenesis: some examples from deep sea In Muller, G. and Friedman, G. M. (Eds.), Carbonate drilling: J. Sediment. Petrol., v. 43, p. 381-390. sedimentology in central Europe: New York (Springer- Deffeyes, K. S., Lucia, F. J., and Weyl, P. K., 1965. Verlag), p. 33-38. Dolomitization of Recent and Plio-Pleistocene sediments Lloyd, R. M. and Hsü, K. J., 1973. Preliminary isotopic by marine evaporite waters on Bonaire, Netherlands investigations of samples from deep-sea drilling in the Antilles. In Pray, L. C. and Murray, R. C. (Eds.), Mediterranean Sea, Leg 13. In Ryan, W. B. F., Hsü, Dolomitiation and limestone diagenesis: Spec. Pub. No. K. J., et al., Initial Reports of the Deep Sea Drilling 13, Soc. Econ. Paleon. Mineral. Project, Volume 13: Washington (U.S. Government Degens, E. T. and Epstein, S., 1964. Oxygen and carbon Printing Office), p. 783-788. isotope ratios in coexisting calcites and dolomites from Marschner, H., 1968. Ca-Mg distribution in carbonates from recent and ancient sediments: Geochim. Cosmochim. the Lower Keuper in NW Germany. In Muller, G. and Acta, v. 28, p. 23-47. Friedman, G. M. (Eds.), Carbonate sedimentology in Edgar, N. T., Saunders, J. B., et al., 1973. Initial Reports central Europe: New York (Springer-Verlag), p. 33-38. of the Deep Sea Drilling Project, Volume 15: McCrea, J. M., 1950. The isotopic chemistry of carbonates Washington (U. S. Government Printing Office). and a paleotemperature scale. J. Chern. Phys., v, 18, p. Engel, A. E. J., Clayton, R. N., and Epstein, S., 1958. 849. Variation in isotopic composition of oxygen and carbon McKinney, C. R., McCrea, J. M., Epstein, S., Allen, H. A., in Leadville limestone and its hydrothermal and and Urey, H. C, 1950. Improvements in mass metamorphic phases: J. Geol., v. 66, p. 374-393. spectrometers for the measurement of small differences Fontes, J-Ch., Letolle, R., Nesteroff, W. D., and Ryan, in isotope abundance ratios: Rev. Sci. Inst., v. 21, p. W. B. F., 1973. Oxygen, carbon, sulfur, and hydrogen 724. stable isotopes in carbonate and sulfate mineral phases Milliman, J. D., Ross, D. A., and Ku, T. L., 1969. of Neogene evaporites, sediments, and in interstitial Precipitation and lithification of deep-sea carbonates in waters, Leg 13. In Ryan, W. B. F., Hsü, K. J., et al., the Red Sea: J. Sediment. Petrol., v. 39, p. 724-736. Initial Reports of the Deep Sea Drilling Project, Volume Nesteroff, W. D., 1973. Mineralogy, petrogrpahy, distri- 13: Washington (U.S. Government Printing Office), p. bution, and origin of the Messinian Mediterranean 788-796. evaporites, Leg 13. In Ryan, W. B. F., Hsü, K. J., et al., Fritz, P. and Smith, D. G. W., 1970. The isotopic Initial Reports of the Deep Sea Drilling Project, Volume composition of secondary dolomites: Geochim. Cosmo- 13: Washington (U. S. Government Printing Office), p. chim. Acta, v. 34, p. 1161-1173. 673-695. Füchtbauer, H. and Goldschmidt, H., 1965. Beziehungen Northrop, D. A. and Clayton, R. N., 1966. Oxygen isotope zwischen calciumgehalt und bildungsbedingungen der fractionations in systems containing dolomite: J. Geol., dolomite: Geol. Rundschau, v. 55, p. 29-40. v. 74, p. 174-196. Gebelein, CD. and Hoffman, P., 1971. Algal origin of Parry, W. T., Reeves, C. C, Jr., and Leach, J. W., 1970. dolomite in interlaminated limestone-dolomite sedimen- Oxygen and carbon isotopic composition of West Texas tary rocks. In Bricker, O. P. (Ed.), Carbonate cements: lake carbonates: Geochim. Cosmochim. Acta, v. 34, p. Baltimore (Johns Hopkins Press). 825-830. Goldsmith, J. R. and Graf, D. L., 1958. Structural and Peterson, M. N. A., Edgar, N. T., et al., 1970. Initial compositional variations in some natural dolomites: J. Reports of the Deep Sea Drilling Project, Volume 2: Geol., v. 66, p. 678-693. Washington (U. S. Government Printing Office). Goldsmith, J. R., Graf, D. L., and Joensuu, O. I., 1955. The Riedel, W. R., Ladd, H. S., Tracey, J. I., Jr., and Bramlette, occurrence of magnesian calcites in nature: Geochim. M. N., 1961. Preliminary drilling phase of the Mohole Cosmochim. Acta, v. 7, p. 212-230. Project II: Summary of coring operations: Am. Assoc. Hayes, D. E., Pimm, A. C, et al., 1973. Initial Reports of Petrol. Geol. Bull., v. 45, p. 1793-1798. the Deep Sea Drilling Project, Volume 14: Washington Ryan, W. B. F., Hsu, K. H., et al., 1973. Initial Reports of (U. S. Government Printing Office). the Deep Sea Drilling Project, Volume 13: Washington Illing, L. V., Wells, A. W., and Taylor, J. C. M., 1965. (U.S. Government Printing Office). Penecontemporary dolomite in the Persian Gulf. In Pray, Sharma, T. and Clayton, R. N., 1965. Measurement of L. C. and Murray, R. C. (Eds.), Dolomitization and 0*8/016 ratios of total oxygen from carbonates. limestone diagenesis: Spec. Pub. No. 13, Soc. Econ. Geochim. Cosmochim. Acta, v. 29, p. 1347. Paleon. Mineral. Spotts, J. H. and Silverman, S. R., 1966. Organic dolomite Lawrence, J. F., in press. Interstitial water studies, Leg 15, from Point Fermin, California: Am. Mineralogist, v. 51, stable oxygen and carbon isotope variations in water, p. 1144-1155.
878